Alcohol dependence (AD) is a common problem with significant health consequences. The discovery that the opioid antagonist naltrexone has efficacy for the treatment of AD was an important breakthrough. However, despite clinical evidence that some patients have an excellent response to naltrexone, the overall treatment effect of naltrexone is moderate and this has discouraged its use. Three markers associated with different aspects of activity of the endogenous opioid system have been identified as having predictive value for naltrexone response: 1) polymorphism of the 5-opioid receptor gene;2) family history of alcoholism;3) severity of craving for alcohol. However, the predictive value of these markers is modest pointing out the need to identify new methods to predict naltrexone response. The main goal of the present proposal is to test the hypothesis that the hedonic response to sweet taste is a novel marker that can help to identify which AD patients have robust responses to naltrexone. The hedonic response to sweet taste (HRST) is a heritable trait associated with genetic risk for alcoholism and one that likely reflects an integrative measure of the endogenous opioid system. Two principal HRST phenotypes have been identified-Sweet Liking (SL) associated with a preference for highly concentrated sweets and Sweet Disliking (SDL) associated with a preference for weaker sweet tastes. In a 12-week, open-label, pilot study of naltrexone (50 mg orally) in 40 AD patients (25 SDL, 15 SL) we evaluated the predictive value of HRST regarding treatment outcome and its interaction with craving for alcohol, another marker related to opioid function. Prior to receiving naltrexone, SL participants took 30% longer to achieve three consecutive abstinent days than SDL individuals (p=.02). During treatment, SDL individuals exhibited 48% days abstinent compared to 30% days abstinent for SL participants (p=0.034). Furthermore, SL individuals required a median of 44 days to achieve two consecutive abstinent days compared to 4 days for SDL participants (Chi[1]=6.88, p<0.01). A combination of response to sweet taste and initial alcohol craving as assessed by the Penn Alcohol Craving Scale (Flannery et al, 1999) was significantly associated with the amount of abstinent days attained during treatment (F[1,36]=13.94, p<0.001)-the combination of the two measures predicted 26.2% of the variance for % days abstinent whereas the SL/SDL phenotype alone predicted 7% of the variance. These findings indicate that HRST may help to differentiate response to naltrexone and that the combination of HRST and craving for alcohol may provide additive information to predict naltrexone response. To confirm and extend these preliminary findings it is essential that a placebo-controlled naltrexone trial be completed. The present proposal will test the above hypothesis utilizing a double-blind, placebo-controlled trial in an adequate number of subjects balanced for SL/SDL phenotype and level of craving. Identifying which patients are more likely to respond to naltrexone would be an important advance towards individualized pharmacotherapy for alcohol dependence. PUBLIC HEALTH RELEVANCE: The identification of predictors of naltrexone response in alcohol dependence is an important research objective to advance clinical care. The pleasurable response to sweet taste represents a probe of the brain opioid reward system that we have shown, along with craving for alcohol, to have predictive value for naltrexone response. In the proposed study, we will conduct a double-blind, placebo-controlled trial to assess how sweet preference and craving for alcohol interact to predict naltrexone response.